Gentamicin serum concentration measurement in children

Discussion

We studied s-gentamicin pre-dose samples in all children who received treatment with gentamicin following admission to four paediatric departments covering the paediatric population in large parts of Northern Norway, Central Norway, Western Norway and Oslo. Among 1,288 treatment episodes, we found a pre-dose s-gentamicin concentration of more than 1.0 mg/L in seven children (0.5 %). In five of the seven children with elevated values, the pre-dose concentration was 1.1–1.5 mg/L. Norwegian paediatric guidelines and international publications recommend extending the dose interval in children with a pre-dose s-gentamicin concentration that is above 1.0 mg/L (6, 12, 17). A small percentage of the children had creatinine levels above the reference range for their age, but none of the children at St. Olav's Hospital or the University Hospital of North Norway developed gentamicin-induced renal failure. Of the patients with an elevated s-gentamicin level in the pre-dose sample, one had a significantly elevated creatinine level, and another had a slightly elevated creatinine level that decreased during treatment. Creatinine levels after treatment initiation were normal in the other children with elevated pre-dose s-gentamicin levels. The majority of children at St. Olav's Hospital and the University Hospital of North Norway were treated with gentamicin for five days or less.

Norwegian guidelines state that for adults with a presumed normal volume of distribution and normal renal function, a pre-dose sample is not indicated for gentamicin treatment of less than three days' duration if no other nephrotoxic drugs are used (19). Children usually have normal renal function that has not yet been affected by degenerative ageing processes. The evidence base for routine measurement of s-gentamicin in children is limited. Nevertheless, standard practice in Norway has been to take a pre-dose sample for the measurement of s-gentamicin immediately before either the second or third dose.

Gentamicin is considered to have a low risk of toxicity in children with no underlying chronic health conditions. However, few studies have examined s-gentamicin concentrations and toxicity in children treated with a once-daily dose of 7 mg/kg. A prospective cohort study from the United States included 79 children aged 1 month to 16 years who underwent a total of 106 episodes of gentamicin treatment (15). Two children (1.9 %) were subsequently diagnosed with hearing impairment in the high-frequency range, and one child (0.9 %) had transient nephrotoxicity. Both children with hearing impairment had also received cisplatin, which is associated with hearing loss, and the child with transient renal impairment had been treated with cyclophosphamide. All three children with reported toxicity had normal s-gentamicin levels according to the Hartford nomogram (20). A study from England included 59 children aged 6 months to 16 years who underwent 113 episodes of gentamicin treatment (21). None of the children had abnormal s-gentamicin levels according to the Hartford nomogram, and thus no dose adjustment was necessary in any of the treatment episodes. No significant nephro- or ototoxicity was observed, and the authors concluded that s-gentamicin measurement is unnecessary for the short-term treatment (less than five days) of children with normal s-creatinine levels who are not receiving other nephrotoxic medications (21). A study from the United States examined s-gentamicin concentrations in 54 children with cancer aged 2–12 years, who were treated for a total of 73 febrile episodes. All received gentamicin 7 mg/kg, which resulted in a mean peak s-gentamicin concentration of 17 mg/L at 30 minutes post-infusion, and a mean s-gentamicin concentration of 0.9 mg/L at 12 hours post-infusion (22). The authors concluded that 7 mg/kg gives rise to an adequate peak concentration, consistent with international guidelines for gentamicin dosage in children (23).

A systematic review from 2021 reported that children with impaired renal function, children with cystic fibrosis who often receive repeated courses of aminoglycosides, and children with cancer who are treated with other ototoxic drugs (especially platinum derivatives) are at risk of developing aminoglycoside-induced hearing loss (16). An extended treatment duration, usually defined as more than a week, and chronic kidney disease are also significant risk factors associated with aminoglycoside-induced nephrotoxicity (10, 24). However, the risk of gentamicin toxicity in other children is very low. A study in adults showed that an initial dose of gentamicin in patients with severe sepsis posed no threat to renal function (25). We have previously shown in a large cohort of neonates treated with gentamicin at a dose of 6 mg/kg that there was no association between s-gentamicin concentration or cumulative gentamicin dose and risk of hearing loss or subclinical kidney damage at school age (26, 27). In our current study, gentamicin was usually administered for three to five days, often with a switch to oral dosing of other antibiotics, particularly in the case of urinary tract infections. We therefore consider the risk of toxicity to be very low.

This study has both strengths and weaknesses. The study population consists of all children treated with gentamicin over a five-year period following admission to four primary hospital departments covering the paediatric population in large areas of Northern Norway, Central Norway, Western Norway and Oslo. Gentamicin treatment and s-gentamicin measurements were performed in accordance with Norwegian paediatric guidelines (17), and the sample is considered representative of Norwegian children requiring treatment with gentamicin. A weakness of the study is that data were collected retrospectively. We cannot rule out the possibility of dosing errors, or of some pre-dose samples not having been taken 24 hours after the first dose and just before the second or third dose. We also lack clinical data from Haukeland University Hospital and Oslo University Hospital, Ullevål, including for the seven children who had elevated gentamicin levels in the pre-dose sample. It is important to be aware that gentamicin is distributed into body water. Overweight is common in Norwegian children (28), and in individuals with significant obesity, dosing by weight may result in too high a dose of gentamicin being administered. However, body weight was not recorded in the present study. A strength of the study is that we have a complete set of gentamicin pre-dose samples from almost 1,300 treatment episodes. No structured follow-up was performed with respect to hearing, but the short treatment durations and low s-gentamicin concentrations indicate that the cumulative gentamicin exposure was low in our cohort. For the majority of children in the study, s-gentamicin levels were below a detection limit of 0.3–0.6 mg/L. While we do not know the exact s-gentamicin levels for these children, other studies have reported that s-gentamicin concentrations are typically below 0.5–1 mg/L after 10–15 hours (15, 22). Gentamicin has a post-antibiotic effect (15, 21, 29), and we found no evidence of therapeutic failure at the two hospitals for which clinical data were available.